Clinical evaluation of spasticity was performed in lower extremities in 35 ambulatory multiple sclerosis patients and compared with the soleus stretch reflex and the Hoffman reflex. There was no relation between the muscle tone score of dorsiflexion of the foot and the biomechanical/electrophysiological parameters. In contrast, the Achilles tendon reflex score was significantly related to the amplitude (rho = 0.411, P < 0.05) and the slope of the stretch reflex (rho = 0.523, P < 0.01). The clinical examination at the ankle joint revealed 33% normal reflex examinations but only 7% normal muscle tone examinations. In contrast, the number of normal examinations of patellar reflex and muscle tone at the knee joint were similar. It is concluded that the muscle tone score overestimates the amount of spasticity because of changes in the non-reflex properties of the spastic extremity and that a reflex score should be used as a clinical measure of spasticity. In addition, biomechanical/electrophysiological evaluation of spasticity at the ankle joint relates to the over-all total muscle tone and reflex scores of lower extremities in this group of MS patients.